Weed Guard: Vision-Guided Weed Management with CoppeliaSim-Driven Simulation for Smart Farming

By 2050, the global population is projected to rise from 8 billion to over 9 billion, necessitating a 60-70% increase in food production. Weed control is a key component in achieving this goal through precision agriculture. Traditional methods, including manual weeding and chemical herbicides, either lack efficiency or pose environmental and economic challenges. This work introduces Weed Guard, a simulation-based weed detection and control framework that integrates MATLAB-driven image processing with robotic behavior simulated in CoppeliaSim. This includes development of an advanced image processing pipeline using contrast stretching, Lab color segmentation, and morphological operations to accurately extract weed patches in dense agricultural fields. Also, implementation of a closed-loop control system to convert weed coordinates into movement commands for the KUKA YouBot, enabling precise and repeatable mechanical actuation based on visual feedback. This approach effectively demonstrates the perception, detection, and path-planning components of an autonomous weed management system. By combining MATLAB image processing capabilities with the CoppeliaSim simulation environment, this work presents a scalable and cost-effective virtual test-bed for precision farming. The proposed system serves as a proof-of-concept for developing real-time, vision-based autonomous weeding robots adaptable to various agricultural settings.